Skip to content

Figures, scripts and data for paper "Spatial resource heterogeneity creates local hotspots of evolutionary potential"

License

Notifications You must be signed in to change notification settings

emilydolson/evo_in_space

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

This repository contains figures, scripts, and data for the paper:

Spatial resource heterogeneity creates local hotspots of evolutionary potential

By Emily Dolson and Charles Ofria

European Conference on Artificial Life, 2017

License: MIT Paper:10.7551/ecal_a_023

Contents:

  • hotspots.tex - LaTeX file for main paper
  • alifeconf.sty - Style file for ECAL
  • Zotero.bib - bibliography file for paper
  • config - contains all files used to generate daa
    • avida - executable used to generate data
    • avida-lineage - executable used to output lineage trajectories
    • avida.cfg - config file for Avida (not changed from default)
    • defaults-heads.org - starting organism (not changed from default)
    • env500XX.cfg - these files indicate how to build the 8 different environments
    • envcontrol.cfg - file to specify the homogeneous control environment
    • events.cfg - tells Avida what data to output when
    • corner_events.cfg - events file that puts starting organism in oppostie corner
    • instset-heads.cfg - tells avida what instruction set to use (not changed from default)
    • lineage_events.cfg - events file for runs where spatial lineage is being output
    • write_events_file.py - python script to make events file that only prints phenotype grid immediately before a new task is going to evolve (assumes that extract_points.py has already been called to determine when each task will evolve)
  • data - contains summary-level data files extracted from raw data (raw data would be too large too include)
    • all_task_locs.csv - the locations where each task first evolved in each run
    • control_diversity_ranks.csv - the percentile of the diversity of the cell where each task first evolved in each replicate of the control condition
    • diversity_ranks.csv - the percentile of the diversity of the cell where each task first evolved in each replicate of the experimental conditions
    • qsub_files - all of the scripts used to submit experiments to MSU's high performance compupting cluster
  • figs - figures
    • 50060_9_paths.png - all lineage paths for task 9 in environment 50060 (not very useful - there are too many)
    • 50065_all_hotspots.png - all hotspots across all environments
    • 9_50013_density.png - kernel density map for task 9 in environment 50013
    • 9_50013_hotspots.png - hotspot locations for task 9 in environment 50013
    • 9_50013_k-hat.png - Ripley's K for task 9 in environment 50013
    • 9_50013_points.png - point locations for task 9 in environment 50013
    • control_9_paths.png - 5 example lineage paths for task 9 for each environment
    • localdiversity.png - qqplot comparing diversity ranks to uniform distribution
    • statsfig.png - diagram outlining statsitics pipeline
    • statsfig_horizontal_1col.png - smaller version of the stats pipeline figure that fits in 1 column
  • scripts - Python and R scripts for processing and analyzing data
    • collect_points.py - grab data from task_locs.csv files for each rep and summarize them into all_task_locs.csv
    • draw_hotspots.py - can be used to make various figures showing points, hotspots, and lineage paths over an environment
    • env500XX.csv - files generated by extract_env_data.py that indicate distance from each cell in each environment to each resource
    • envcontrol.csv - like above, except it's all 0s
    • extract_diversity.py - generate diversity_ranks.csv file from raw phenotype grids
    • extract_env_data.py - generate csv files indicating the distance from each cell in an environment to each resource
    • extract_grids.py - grab phenotype grids from right before each task evolves out of tar archives
    • extract_paths.csv - convert sequence of ints output by avida-lineage into actual sequences of coordinates and environments that a lineage passed through
    • extract_points.csv - loop through phenotype grids to figure out when each task first appears (generates task_locs.csv)
    • kriging.R - R script that handles all of the regression that we used in an attempt to figure out what causes hotspots
    • make_diversity_maps.csv - makes heat maps of local diversity for updates immediately before tasks evolved
    • make_hotspot_figure.py - variant on draw_hotspots.py, used to make multi-panel hotspot figure in paper
    • make_path_figure.py - variant on draw_hotspots.py used to make multipanel path figure in paper

About

Figures, scripts and data for paper "Spatial resource heterogeneity creates local hotspots of evolutionary potential"

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published